Leave a comment

It's that time of year again...do you have a CRO plan for the holidays?

  • SJ

    Sebastian Johansson

    over 3 years ago #

    Good article. Ive always felt less enthusiastic about ab-tests during spikes, like a newspaper article mentioning your company or Christmas.

    The visitors and their behaviour wont be representative for the users of the rest of the year. And its over fast.

    Although, holiday users might behave the same the next year.

    But for high revenue companies, it makes sense to develop a special design optimized for just Christmas etc based on results from AB-tests during these just these times.

    Maybe even having a special design for visitors coming from newspaper articles. So if one notice a high spike from New York Times, and have previously concluded through ab-tests that those kind of visitors are less educated/less eager about your product and a more explanatory design converts them better, the design can be switched on for just that set of visitors.

    What saas tool would be the easiest to use to show a certain version of your website based on where the visitor come from? And without upsetting google and your seo rankings.

    • ES

      Edward Stephens

      over 3 years ago #

      You are absolutely right Sebastian, this kind of customisation is crucial. Anuj posted a video by Bill Gurley which dived into this which was fascinating.

      It's been interesting to see the effect of Black Friday and Cyber Monday in the UK. It was a nothing event in 2012/13 but in the last 2 years, it's really impacted/changed the consumer shopping behaviour in the lead up to Christmas.

      Offline retailers begun marketing Christmas from November 1st this year. Then accounting for the January sales this period has been stretched to almost 2.5 months.

      • SJ

        Sebastian Johansson

        over 3 years ago #

        Great. Will check it out.

        Im guessing you will see different behaviours during that 2.5 period as well.

        A header saying "You will get it before Christmas!" will probably convert better last week of Dec than in Nov/Jan. Jan visitors might be looking for some special deals etc etc.

        So I dont feel you have 2.5 months to test "the Christmas design".

        But yeah. I guess one take this whole customisation as deep as one wants. Like visitors on Fridays vs Mondays. Visitors from Manchester on a Friday vs visitors from South London on a Friday.

        Problem for most companies is probably simply the lack of enough conversions to set up all the AB-tests they want.

        I would wanna see an ab-test between companies that only optimize for average customers throughout the year vs companies that try to optimize especially for Christmas shoppers as well.

        Companies that do test for Christmas loose 2.5 months worth of testing each year. That period might create more bottom line revenue to by using it to do tests that will have an effect all year round.

        I wonder if testing different designs for Christmas is just part of a social narrative of the cro-world (because it sounds logical) or if there are actually solid evidence.

  • ES

    Edward Stephens

    over 3 years ago #

    Well I think the strangest thing about it was that a US consumer habit swept through the UK in under 24 months.

    Now as you say, the marketing/E-comms teams will be collecting the data and incentivising/driving a new wave of purchasing behaviour.

    I'm working with a number of fashion tech start-ups and I think as you pointed out rather than a single 'Christmas peak' we will start to see multiple spikes.

    I think to build on your original point I would like to know what analytcis/tools people are typically relying on. Also, how they are using mobile discovery to drive Christmas purchasing decisions or whether they are driving customers to buy directly via M-commerce.

    • SJ

      Sebastian Johansson

      over 3 years ago #

      Mm, interesting.

      Think I read that companies grow and die faster than ever before (I could be wrong).

      And I guess its the same with culture. The evolution of both companies and culture, like Black Friday and Cyber Monday, are speeding up rapidly.

      I guess as a growth hacker one needs to focus on developing skills that are more likely to stand the test of time. For instance, frameworks for growth/testing, building growth teams etc.

      All the tools one use today will probably be obsolete soon. But one needs them today, so its a balance I guess.

  • DL

    Daniel Louis

    over 3 years ago #

    Great article! Thanks for sharing

Join over 70,000 growth pros from companies like Uber, Pinterest & Twitter

Get Weekly Top Posts
High five! You’re in.
SHARE
46
46